Unmanned aerial vehicle flight navigation control method and system based on VR head-mounted display

Abstract

The invention provides an unmanned aerial vehicle flight navigation control method and system based on a VR head-mounted display, and relates to the technical field of unmanned aerial vehicle control. Dynamic path optimization and intelligent standby landing point management are adopted, so that the optimal standby landing point can be quickly screened and sorted in emergency situations such as insufficient electric quantity; it is ensured that the unmanned aerial vehicle can land safely within the minimum energy consumption range; besides, by analyzing the combination of historical power consumption data and real-time data, the system can intelligently adjust a flight strategy, the flexibility and adaptability of unmanned aerial vehicle operation are improved, and risks in emergency situations are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV control, and in particular to a UAV flight navigation control method and system based on a VR headset. Background Art

[0002] In the field of UAV control, the introduction of VR technology has brought new innovation points to traditional remote control operations. Traditional UAV remote control relies on two-dimensional screen display. The operator controls the UAV through the limited perspective provided by the ground control station or handheld device, and is often restricted by visual perception, making it difficult to timely and accurately perceive the changes in the environment around the UAV. The introduction of VR headsets can enhance the operator's intuitive understanding and reaction speed of the UAV flight state through three-dimensional stereoscopic views and real-time environment perception. However, current research and practical applications in this field still face numerous technical challenges.

[0003] In the prior art, the publication number is CN109407681A, and the name is a UAV flight control method, a UAV flight control device, a UAV, and a storage medium. The method includes: controlling the transmission signal emitted by the radar to the ground during the flight of the UAV and receiving the echo signal after the transmission signal is reflected; determining the target measurement distance from the UAV to the ground or the plant surface according to the transmission signal and the echo signal; judging whether the target measurement distance is within a preset safety distance range; if not, adjusting the flight height of the UAV according to the target measurement distance and the preset safety distance range, so that the UAV flies in imitation of the plant surface or the ground according to the adjusted flight height. The accurate distance from the UAV to the plant surface or the ground can be obtained through the transmission signal and the echo signal of the radar, so that the UAV can accurately adjust its flight height and realize accurate ground or plant surface imitation flight of the UAV.

[0004] Existing UAV control methods still have deficiencies in precise real-time navigation and path management in complex environments: traditional control systems lack efficient and flexible emergency landing planning capabilities. In the face of emergencies (such as insufficient power or weather changes), the operator needs to make a quick decision to select a suitable landing point to ensure the safe landing of the UAV. However, existing technologies usually rely on pre-planned static landing points and lack the functions of dynamic evaluation and real-time adjustment; in addition, in the control system combined with VR headsets, there are problems such as untimely processing of path visual image data, insufficient accuracy, and slow environmental feedback. This results in a lack of reliability and accuracy in the response of existing systems when quickly processing environmental information and making reasonable decisions.

[0005] The above information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and therefore it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The object of the present invention is to provide a method and system for controlling the flight navigation of an unmanned aerial vehicle based on a VR headset, so as to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for controlling the flight navigation of an unmanned aerial vehicle based on a VR headset, the specific steps include:

[0009] Step S1: Obtain the flight navigation path and path visual image data of the unmanned aerial vehicle in the current time period;

[0010] And divide the flight path into multiple continuous sections. At the same time, obtain the historical power consumption data of the unmanned aerial vehicle in each section;

[0011] Step S2: Receive the flight navigation path and path visual image data for analysis and processing, mark multiple emergency landing points on the flight navigation path, and determine the safety landing area of each emergency landing point;

[0012] Step S3: If the unmanned aerial vehicle is in a low-power alarm state, first determine the remaining flight distance of the unmanned aerial vehicle based on the historical power consumption data of each section, and then circle a circular screening range with the remaining flight distance as the radius, and initially screen out a set of reachable landing points from multiple emergency landing points;

[0013] Step S4: Perform secondary analysis on the path visual image data of the surrounding environment of each emergency landing point in the set of landing points to generate a landing reliability index for these selected emergency landing points, and this landing reliability index is used to rank the priority of each emergency landing point in the set of landing points.

[0014] A system for controlling the flight navigation of an unmanned aerial vehicle based on a VR headset, the system is used to execute the method for controlling the flight navigation of an unmanned aerial vehicle based on a VR headset, including:

[0015] Data acquisition module: used to obtain the flight navigation path and path visual image data of the unmanned aerial vehicle in the current time period;

[0016] And divide the flight path into multiple continuous sections. At the same time, obtain the historical power consumption data of the unmanned aerial vehicle in each section;

[0017] Emergency landing point marking module: used to receive the flight navigation path and path visual image data for analysis and processing, mark multiple emergency landing points on the flight navigation path, and determine the safety landing area of each emergency landing point;

[0018] Screening module: When the drone is in a low - power alarm state, it first determines the remaining flight distance of the drone based on the historical power consumption data of each section of the road, and then circles a circular screening range with this remaining flight distance as the radius to preliminarily screen out a set of reachable emergency landing points from multiple emergency landing points;

[0019] Priority sorting module: It performs a secondary analysis on the path visual image data of the surrounding environment of each emergency landing point in the set of landing points to generate a landing reliability index for these selected emergency landing points, and this landing reliability index is used to sort the priority of each emergency landing point in the set of landing points.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0021] 1. Enhanced environmental perception and interaction experience: Utilizing the immersive perspective and real - time interaction characteristics of the VR headset, the operator can obtain more intuitive three - dimensional navigation information and environmental feedback; this immersive experience enables the operator to have higher precision during the decision - making process, effectively improving the safety and control efficiency during the flight of the drone;

[0022] 2. Dynamic flight path and landing management: Compared with the traditional static emergency landing point planning method, the present invention provides efficient dynamic path optimization and emergency landing point selection through real - time analysis of the drone's flight path and energy consumption; especially when dealing with insufficient power or other emergencies, the system can quickly screen and sort the emergency landing points to ensure that the drone can land safely within the minimum energy consumption range, reducing the risks caused by unexpected situations;

[0023] 3. Intelligent energy consumption assessment and adjustment mechanism: By analyzing the historical power consumption data of the drone and comparing the real - time flight data, the new method can accurately evaluate the flight energy consumption and automatically generate a correction coefficient to optimize the flight strategy; this function ensures that the drone has higher flexibility and adaptability in path planning and emergency landing selection, effectively coping with the changing flight environment. Brief description of the drawings

[0024] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0025] Figure 2 It is a block diagram of the overall system module of the present invention. Detailed implementation manners

[0026] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with specific embodiments.

[0027] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the art to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative position relationships, and when the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0028] Embodiment 1:

[0029] Please refer to Figure 1 , the present invention provides a technical solution:

[0030] A method for controlling the flight navigation of a drone based on a VR headset, which uses the VR headset for flight navigation control. The specific steps include:

[0031] Step S1: Obtain the flight navigation path and path visual image data of the drone in the current time period;

[0032] And divide the flight path into multiple consecutive sections. At the same time, obtain the historical power consumption data of the drone under each section;

[0033] Further, denote the multiple consecutive sections as {1, 2,..., i,..., N}, where i represents the index of the i-th section and N is the total number of divided sections;

[0034] Step S2: Receive the flight navigation path and path visual image data for analysis and processing to mark multiple emergency landing points on the flight navigation path and determine the safety landing area area of each emergency landing point;

[0035] Further, the analysis and processing of the path visual image data include: after image recognition, obtain the terrain data, obstacle data and movement data of moving objects under each section;

[0036] Analyze the terrain data, obstacle data and movement data of moving objects to generate a first terrain flatness coefficient, a first obstacle avoidance coefficient and a first dynamic safety coefficient respectively;

[0037] Based on the path segments divided in step S1, map the flight path to a one-dimensional time axis according to the segment index. In each segment, clarify the flight direction, speed, and path inclination angle of the UAV.

[0038] Perform data preprocessing on the path visual images:

[0039] Segment the visual image data collected by the UAV camera according to the time axis and correspond them one by one with the path segments;

[0040] Crop each segment of the image separately to ensure that each segment of image data only covers the flight path and its surrounding preset area.

[0041] Use image denoising techniques (such as Gaussian blur or bilateral filtering) to filter out environmental interference to improve the accuracy and efficiency of subsequent feature extraction.

[0042] Perform low-quality data elimination:

[0043] Data quality determination: If the visual image of a certain segment is blurred (such as motion blur or sensor signal loss), directly eliminate the data of the corresponding segment and mark it as a "data missing segment".

[0044] If there are 3 or more consecutive "data missing segments", immediately trigger the warning mechanism and prompt the operator.

[0045] Scene feature extraction:

[0046] After data preprocessing, based on multiple consecutive segments {1, 2,..., i,..., N}, analyze the path visual image data of each segment based on different feature dimensions, extract several important feature indicators, and form multi-dimensional scene data:

[0047] Terrain data extraction:

[0048] Use the semantic segmentation algorithm of deep learning (in this embodiment, use the convolutional neural network CNN, such as the U-Net architecture) to perform pixel-by-pixel classification on the image, extract different terrain regions to distinguish flat ground, slopes, water surfaces, and sandy lands.

[0049] Layer the visual images according to the class mask (Mask), and count the proportion of flat areas in the candidate area for the alternate landing point. Specifically, quantify the proportion of flat areas as the "first terrain flatness coefficient"; focus on screening out flatness areas with ≥ 80% as potential alternate landing candidate areas.

[0050] Obstacle data extraction:

[0051] Use the object detection algorithm (such as YOLO or Faster R-CNN) to identify the types of obstacles (buildings, vegetation, wires) and their positions from the image.

[0052] Obtained through stereo vision or depth cameras to calculate the shortest distance between obstacles and the current position of the drone.

[0053] Extraction of movement data of moving objects:

[0054] Using the optical flow algorithm (Optical Flow) or object tracking algorithms (such as DeepSORT) to analyze the trajectory, speed, and direction of moving objects in the image.

[0055] Statistically count the number of dynamic objects, movement intensity, and direction consistency within each candidate alternate landing point area, and quantitatively generate the "first dynamic safety factor".

[0056] Comprehensively mark the emergency alternate landing points:

[0057] Based on the results of scene feature extraction, complete the marking of emergency alternate landing points through the following steps:

[0058] Candidate area division:

[0059] Based on terrain data, filter out areas above a specified flatness threshold as candidate alternate landing areas.

[0060] Exclude areas with obstacle density higher than the set safety threshold.

[0061] Exclude areas with the number of dynamic objects higher than the critical value (in this embodiment, 5 dynamic objects / ㎡).

[0062] Define the first terrain flatness coefficient of the i-th section as PTD i ; The calculation formula is as follows:

[0063]

[0064] Where PTD i has a numerical range of [0, 1], and the closer it is to 1, the flatter the terrain;

[0065] A flat,i is the total area of the flat area in the image corresponding to section i; A flat,i is calculated through the flat area mask generated by the semantic segmentation algorithm and converted to physical area using pixel counting;

[0066] A total,i is the total area covered by the image of section i;

[0067] Use the semantic segmentation algorithm to classify each pixel of the visual image to generate a multi-class mask (such as flat ground, slope, water surface);

[0068] Calculate the cumulative pixel area of the flat category to form A flat , and all valid pixels within the image range form A total;

[0069] PTD i The higher the value of

[0070] Define the first obstacle avoidance coefficient of the i-th section as BRN i ; The calculation formula is as follows:

[0071]

[0072] where BRN i The value range is [0, 1], and the closer BRN i is to 1, the smaller the threat of the obstacle;

[0073] D avg,i is the average shortest distance between the obstacle and the current position of the UAV in section i;

[0074] D max is the preset maximum safety distance threshold; if the distance between the obstacle and the UAV is greater than this value, it is considered non-threatening;

[0075] ρ i is the density of obstacles within section i; and the density represents the number of obstacles per unit area;

[0076] ρ max is the specified maximum tolerable obstacle density; the area exceeding this density will be directly excluded as an alternate landing point; ρ max is determined based on relevant technical manuals or the expert experience knowledge of UAVs;

[0077] b1 and b2 are the distance weight and the density weight respectively, following b1 + b2 = 1; and the values of b1 and b2 are in the interval (0, 1); if the distance weight is more important, let b1 = 0.7 and b2 = 0.3;

[0078] Use the object detection algorithm to identify obstacles, including but not limited to trees, buildings, wires, etc., and calculate their position distances;

[0079] Based on the distribution density and distance of the obstacles, comprehensively consider the threat posed by the obstacles to the alternate landing safety. The higher the value, the smaller the interference of the obstacles;

[0080] BRN i The closer the value is to 1, the smaller the impact of the obstacle on the alternate landing point, providing a basis for the selection of a relatively safe alternate landing point.

[0081] Define the first dynamic safety coefficient of the i-th section as DAQi; the calculation formula is as follows:

[0082]

[0083] Among them, the value range of DAQi is [0, 1]. The closer DAQi is to 1, the safer the dynamic environment is;

[0084] n i is the total number of dynamic objects in the i-th section;

[0085] n max is the threshold of the maximum number of allowed dynamic objects; exceeding this value, the area is considered dynamically unsafe;

[0086] v avg,i is the average speed of dynamic objects in the i-th section;

[0087] v max is the specified maximum safe speed threshold; dynamic objects exceeding this speed will significantly increase the risk of diversion; v max is determined based on relevant technical manuals or the expert experience knowledge of drones;

[0088] θ i is the motion direction deviation of dynamic objects in the i-th section;

[0089] θ max is the maximum effective value of the motion direction deviation; exceeding this direction change range, the dynamic environment is considered too complex;

[0090] c1, c2, and c3 are the weights of the number, speed, and direction of dynamic objects respectively, satisfying c1 + c2 + c3 = 1; the values of c1, c2, and c3 are in the interval (0, 1);

[0091] The number, speed, and motion direction of dynamic objects are statistically calculated in real time through the optical flow algorithm or the target tracking algorithm;

[0092] Combined with the distribution characteristics of dynamic objects within the field of view, the uncertainty of the dynamic environment is quantitatively evaluated;

[0093] The closer the value of DAQi is to 1, the safer the dynamic environment is, and it is suitable for the drone to select as a diversion point.

[0094] Calculate the first comprehensive score for each eligible candidate area and sort them:

[0095] ZHF i = a1 × PTD i + a2 × BRN i + a3 × DAQ i

[0096] Among them, ZHF i is the first comprehensive score of the i-th section in {1, 2,..., i,..., N}, PTD iis the first terrain flatness coefficient of the i-th section; BRN i is the first obstacle avoidance coefficient of the i-th section; DAQ i is the first dynamic safety coefficient of the i-th section;

[0097] a1, a2, and a3 are the weight coefficients of each index respectively. The values of a1, a2, and a3 are in the interval (0, 1); a1 + a2 + a3 = 1; The specific values of a1, a2, and a3 are determined by the entropy weight method and the fuzzy analytic hierarchy process (FAHP) for the corresponding weights;

[0098] Set the high score threshold of the first comprehensive score as ZHF th ; ZHF th It is determined based on relevant technical manuals or the expert experience knowledge of drones;

[0099] Regard the sections that meet ZHF i ≥ZHF th as the emergency landing points; And regard the total area of the flat areas at the positions where each section is located as the safe landing area; And store their coordinate data.

[0100] If the integrals of multiple candidate areas are similar, preferentially select the area with a shorter flight path distance as the landing point.

[0101] In the VR headset interface, use the visual UI to present the positions of the landing points and the first comprehensive score, and at the same time mark the key information of the surrounding environment. The key information includes the first terrain flatness coefficient, the first obstacle avoidance coefficient, and the first dynamic safety coefficient.

[0102] After marking, perform reliability verification according to the data model of the landing point:

[0103] 1. Confirm the boundary of the landing space:

[0104] Simulate the landing trajectory of the drone (including the descent angle, required space, etc.), and perform dynamic analysis on the marked points and the intervals of the surrounding distances from obstacles.

[0105] If the space of the landing point is insufficient, automatically downgrade this marked point to the second choice and re-rank it.

[0106] 2. Real-time dynamic monitoring of the environment:

[0107] Continuously monitor the information of dynamic objects around the landing point to ensure that the safe zone of the marked point will not fail due to changes in the dynamic environment during the flight process.

[0108] Step S3: If the UAV is in a power shortage alarm state, first determine the remaining flight distance of the UAV based on the historical power consumption data of each road section, then define a circular screening range with the remaining flight distance as the radius, and preliminarily screen out a set of accessible emergency landing points from multiple emergency landing points;

[0109] Furthermore, the battery management system (BMS) on the drone is used to obtain the current power data in real time.

[0110] Ensure that the power collection frequency is sufficient to meet the real-time flight path adjustment requirements (at least once per second in this embodiment).

[0111] Once the power level is lower than a preset threshold, which in this embodiment is 20% remaining, the system automatically triggers a power shortage alarm state.

[0112] At the same time, visual and auditory alarms are issued on the drone operation terminal and monitoring system interface to alert the operator.

[0113] The historical power consumption data analysis is as follows:

[0114] 1. Data preparation:

[0115] Extract historical power consumption data of each road section from the cloud or local database.

[0116] The data includes: average power consumption, maximum power consumption, route length, and flight environment conditions (such as wind speed, temperature, etc.).

[0117] 2. Data cleaning and feature extraction:

[0118] Pre-process historical data to remove outliers (such as sudden increases in power consumption caused by extreme weather conditions).

[0119] Calculate the average power consumption per unit distance E for each road section avg,i , the formula is as follows:

[0120]

[0121] Where E total,i is the total power consumption of road section i, L i is the length of the road section;

[0122] Define the remaining flight distance as D remain , the calculation formula is as follows:

[0123]

[0124] Where E current is the current remaining power, E avgis the average power consumption per unit distance for all sections;

[0125] If D remain <L i , that is, the remaining power is not enough to fly through the entire section, then the flight path needs to be adjusted, and the number of emergency landing points that can be reached with the current remaining power is determined as M;

[0126] Prioritize the shortest path section that can be completed with the current power, or find the emergency landing point closest to the end of the endurance.

[0127] Denote the set of emergency landing points as {1, 2,..., j,..., M}, where j represents the index of the j-th emergency landing point, M is the total number of selected emergency landing points, and M ≤ N.

[0128] Step S4: Perform secondary analysis on the path visual image data of the surrounding environment of each emergency landing point in the set of emergency landing points to generate the landing reliability index for these selected emergency landing points, and this landing reliability index is used to rank the priority of each emergency landing point in the set of emergency landing points;

[0129] Furthermore, the generation of the landing reliability index specifically includes:

[0130] For the path visual image data of the surrounding environment of the emergency landing point: Create an annular extended area around the circular screening range, and select a specific area consistent with the direction of the drone in the annular extended area;

[0131] Definition of the annular extended area: Taking each emergency landing point as the center, set a circular area with a fixed radius r1, and r1 is greater than the longest width of the area where each emergency landing point is located;

[0132] On the basis of r1, expand an annular area with a fixed width wd outward to form an annular extended area with a radius of r1 + wd;

[0133] Data collection and preprocessing: Extract the path visual image data taken by the drone in this annular extended area; Use a screening algorithm based on direction perception to select the image data area consistent with the current flight direction of the drone to ensure the direction consistency of data collection.

[0134] After performing image recognition processing on the path visual image data of this specific area, sequentially obtain the second terrain flatness coefficient, the second obstacle avoidance coefficient, and the second dynamic safety coefficient for describing the landing situation of the selected area;

[0135] Define the second terrain flatness coefficient, the second obstacle avoidance coefficient, and the second dynamic safety coefficient as PTD j , BRN j and DAQ j ; PTDj , BRN j and DAQ j are calculated in the same way as the corresponding PTD i , BRN i and DAQ i and will not be elaborated here;

[0136] Analyze the second terrain flatness coefficient, the second obstacle avoidance coefficient, and the second dynamic safety coefficient to calculate and generate the alternate landing reliability index; and define the alternate landing reliability index of the j-th emergency alternate landing point as RI j , and the calculation formula is as follows:

[0137]

[0138] where μ1 is a correction factor used to adjust the magnitude of the change in the value of (e1×PTD j +e2×BRN j +e3×DAQ j ), and the value of μ1 is in the interval (0.2, 0.85); e1, e2, and e3 are the weight coefficients of the corresponding parameters, and the values of e1, e2, and e3 are all in the interval (0, 1), and e1 + e2 + e3 = 1; e1, e2, and e3 are determined for the corresponding weights through the entropy weight method and the fuzzy analytic hierarchy process (FAHP);

[0139] Set the effective value range of RI j to the interval (0, 1); the closer RI j is to 1, the higher the priority ranking of the j-th emergency alternate landing point.

[0140] Step S5: According to the priority ranking result, select the two emergency alternate landing points with the highest priority from the set of alternate landing points, and preferentially control the drone to fly to the one with the farthest distance among the selected emergency alternate landing points, and collect the power consumption data of the drone at the distance of a single flight section. Compare and analyze this power consumption data with the historical power consumption data to generate a correction coefficient, which is used to determine whether to turn to another selected emergency alternate landing point with a shorter distance.

[0141] Furthermore, define the two emergency alternate landing points with the highest priority as j1 and j2, and both j1 and j2 are included in {1, 2,..., j,..., M}; where j1 represents the emergency alternate landing point with the farthest distance from the drone;

[0142] During the process of the drone flying to the j1-th emergency alternate landing point, represent the power consumption data at the distance of a single flight section as HDE j1 ;

[0143] Define the corresponding correction coefficient as JZ j1, the calculation formula is as follows:

[0144]

[0145] Among them, E avg,j1 is the power consumption data for flying a single section distance in the historical power consumption data;

[0146] If JZ j1 > 1, it means that the power consumption of the UAV in the current environment is greater than the corresponding historical power consumption, and it is necessary to turn to another selected emergency landing point with a shorter distance.

[0147] Embodiment 2:

[0148] Please refer to Figure 2 , a UAV flight navigation control system based on a VR headset, the system is used to execute the UAV flight navigation control method based on the VR headset, including:

[0149] Data acquisition module: used to acquire the flight navigation path and path visual image data of the UAV in the current time period;

[0150] And divide the flight path into multiple continuous sections. At the same time, acquire the historical power consumption data of the UAV in each section;

[0151] Emergency landing point marking module: used to receive the flight navigation path and path visual image data for analysis and processing, mark multiple emergency landing points on the flight navigation path, and determine the safe landing area area of each emergency landing point;

[0152] Screening module: used to, if the UAV is in a power shortage alarm state, first determine the remaining flight distance of the UAV based on the historical power consumption data of each section, and then circle a circular screening range with the remaining flight distance as the radius, and initially screen out a set of reachable landing points from multiple emergency landing points;

[0153] Priority sorting module: perform secondary analysis on the path visual image data of the surrounding environment of each emergency landing point in the set of landing points to generate a landing reliability index for these selected emergency landing points, and this landing reliability index is used to perform priority sorting on each emergency landing point in the set of landing points.

[0154] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disc of a computer, etc., including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0155] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0156] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0157] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which a program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or, as necessary, other suitable processing, and then stored in a computer memory.

[0158] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

[0159] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for controlling the flight navigation of an unmanned aerial vehicle based on a VR head display, which uses a VR head display to perform flight navigation control, characterized in that: The specific steps include: Step S1: Obtain the flight navigation path and path visual image data of the UAV in the current time period; The flight path is divided into multiple continuous sections, and the historical power consumption data of the drone in each section is obtained; Step S2: receiving the flight navigation path and the path visual image data for analysis and processing, so as to mark a plurality of emergency alternate landing points on the flight navigation path, and determine the safe alternate landing area of ​​each emergency alternate landing point; Step S3: If the UAV is in a power shortage alarm state, first determine the remaining flight distance of the UAV based on the historical power consumption data of each road section, then define a circular screening range with the remaining flight distance as the radius, and preliminarily screen out a set of accessible emergency landing points from multiple emergency landing points; Step S4: Perform secondary analysis on the path visual image data of the surrounding environment of each emergency alternate landing point in the alternate landing point set to generate an alternate landing reliability index for these selected emergency alternate landing points, and the alternate landing reliability index is used to prioritize each emergency alternate landing point in the alternate landing point set.

2. The method for controlling the flight navigation of an unmanned aerial vehicle based on a VR head display according to claim 1, characterized in that: Multiple continuous road segments are recorded as {1, 2, ..., i, ..., N}, where i represents the index of the i-th road segment and N is the total number of divided road segments; Analyze and process the path visual image data, including: after image recognition, obtain the terrain data, obstacle data and movement data of moving objects under each road section; Analyzing terrain data, obstacle data, and movement data of the moving object to generate a first terrain flatness coefficient, a first obstacle avoidance coefficient, and a first dynamic safety factor, respectively; Define the first terrain flatness coefficient of the i-th road section as PTD i ; PTD i The value range is [0,1], and the closer it is to 1, the flatter the terrain; Define the first obstacle avoidance coefficient of the i-th road segment as BRN i ; BRN i The value range is [0,1], BRN i The closer it is to 1, the smaller the obstacle threat; Define the first dynamic safety factor of the i-th section as DAQ i ; DAQ i The value range is [0,1], DAQ i The closer it is to 1, the safer the dynamic environment is; Calculate the first comprehensive score: ZF i =a1×PTD i +a2×BRN i +a3×DAQ i Among them, ZHF i is the first comprehensive score of the i-th road segment in {1,2,…,i,…,N}, PTD i is the first terrain flatness coefficient of the i-th road section; BRN i is the first obstacle avoidance coefficient of the i-th road section; DAQ i is the first dynamic safety factor of the ith road section; a1, a2 and a3 are weight coefficients of each indicator, and the values ​​of a1, a2 and a3 are in the interval (0,1); a1+a2+a3=1; the high score threshold of the first comprehensive score is set to ZHF th ; ZHF i ≥ZHF th The road section is used as an emergency landing point.

3. The method for controlling the flight navigation of an unmanned aerial vehicle based on a VR head display according to claim 2, characterized in that: Calculate the average power consumption per unit distance E for each road section avg,i , the formula is as follows: Where E total,i is the total power consumption of road section i, L i is the length of the road section; Define the remaining flight distance as D remain , the calculation formula is as follows: Where E current is the current remaining power, E avg is the average power consumption per unit distance of all road sections; If D remain <L i , that is, the remaining power is not enough to fly the entire route, then the flight path needs to be adjusted to the emergency landing point, and the number of emergency landing points that can be reached with the current remaining power is determined as M; The set of alternate landing points is recorded as {1, 2, ..., j, ..., M}, where j represents the index of the jth emergency alternate landing point, M is the total number of selected emergency alternate landing points, and M≤N.

4. The method for controlling the flight navigation of an unmanned aerial vehicle based on a VR head display according to claim 1, characterized in that: The generation of the alternate landing reliability index includes: For the path visual image data of the environment around the emergency landing point: create a circular expansion area around the circular screening range, and select a specific area in the circular expansion area that is consistent with the direction of the drone; Definition of the annular expansion area: With each emergency landing point as the center, a circular area with a fixed radius r1 is set, and r1 is greater than the longest width of the area where each emergency landing point is located; Based on r1, a ring area with a fixed width wd is expanded outward to form a ring expansion area with a range radius of r1+wd; After performing image recognition processing on the path visual image data of the specific area, a second terrain flatness coefficient, a second obstacle avoidance coefficient, and a second dynamic safety factor for describing the alternate landing situation of the selected area are obtained in sequence; The second terrain flatness coefficient, the second obstacle avoidance coefficient and the second dynamic safety factor are defined as PTD respectively. j , BRN j and DAQ j ; The second terrain flatness coefficient, the second obstacle avoidance coefficient and the second dynamic safety factor are analyzed to calculate the alternate landing reliability index; and the alternate landing reliability index of the jth emergency alternate landing point is defined as RI j , the calculation formula is as follows: Where μ1 is the correction factor for (e1×PTD j +e2×BRN j +e3×DAQ j ) is adjusted by the numerical change range, and the value of μ1 is in the interval (0.2, 0.85); e1, e2 and e3 are the weight coefficients of the corresponding parameters, and the values ​​of e1, e2 and e3 are all in the interval (0, 1), e1 + e2 + e3 = 1; Setting RI j The valid range of RI is the interval (0,1); j The closer it is to 1, the higher the priority of the j-th emergency landing point.

5. The method for controlling the flight navigation of a UAV based on a VR head display according to claim 4, characterized in that: According to the priority sorting result in step S4, the two emergency landing points with the highest priority are selected from the set of emergency landing points, and the UAV is preferentially controlled to fly to the one with the longest distance among the selected emergency landing points. The power consumption data of the UAV at a single flight section distance is collected, and the power consumption data is compared and analyzed with the historical power consumption data to generate a correction coefficient. The correction coefficient is used to determine whether it is necessary to turn to another selected emergency landing point that is closer.

6. The method for controlling the flight navigation of a UAV based on a VR head display according to claim 5, characterized in that: The two emergency landing points with the highest priority are defined as j1 and j2, and both j1 and j2 are included in {1, 2, ..., j, ..., M}; j1 is characterized as the emergency landing point farthest from the UAV; When the UAV flies to the j1th emergency landing point, the power consumption data of a single flight section is represented as HDE j1 ; Define the corresponding correction coefficient as JZ j1 , the calculation formula is as follows: Among them, E avg,j1 It is the power consumption data of a single flight section in the historical power consumption data; If JZ j1 When it is greater than 1, it means that the power consumption of the UAV in the current environment is greater than the corresponding historical power consumption, and it needs to turn to another selected emergency landing point that is closer.

7. A UAV flight navigation control system based on VR head display, characterized in that: The system is used to execute the unmanned aerial vehicle flight navigation control method based on a VR head display according to any one of claims 1 to 6, comprising: Data acquisition module: used to obtain the flight navigation path and path visual image data of the drone in the current time period; The flight path is divided into multiple continuous sections, and the historical power consumption data of the drone in each section is obtained; Emergency alternate landing point marking module: used to receive the flight navigation path and path visual image data for analysis and processing, so as to mark multiple emergency alternate landing points on the flight navigation path and determine the safe alternate landing area of ​​each emergency alternate landing point; Screening module: if the drone is in a power shortage alarm state, first determine the remaining flight distance of the drone based on the historical power consumption data of each section, then define a circular screening range with the remaining flight distance as the radius, and preliminarily screen out a set of accessible alternate landing points from multiple emergency alternate landing points; Priority sorting module: Perform secondary analysis on the path visual image data of the surrounding environment of each emergency alternate landing point in the alternate landing point set to generate the alternate landing reliability index of these selected emergency alternate landing points. The alternate landing reliability index is used to prioritize each emergency alternate landing point in the alternate landing point set.

Images